deepgeek/CosyVoice
2
0
Fork 0
mirror of https://github.com/FunAudioLLM/CosyVoice.git synced 2026-02-05 01:49:25 +08:00
Code Issues Packages Projects Releases Wiki Activity

update fastapi

Browse Source
This commit is contained in:
lyuxiang.lx
2024-09-05 14:09:40 +08:00
parent 2ce724045b
commit 7555afb90a
6 changed files with 131 additions and 148 deletions
Download Patch File Download Diff File Expand all files Collapse all files

9
cosyvoice/cli/cosyvoice.py
View File

@@ -13,6 +13,7 @@
# limitations under the License.
import os
import time
from tqdm import tqdm
from hyperpyyaml import load_hyperpyyaml
from modelscope import snapshot_download
from cosyvoice.cli.frontend import CosyVoiceFrontEnd
@@ -52,7 +53,7 @@ class CosyVoice:
return spks
def inference_sft(self, tts_text, spk_id, stream=False):
for i in self.frontend.text_normalize(tts_text, split=True):
for i in tqdm(self.frontend.text_normalize(tts_text, split=True)):
model_input = self.frontend.frontend_sft(i, spk_id)
start_time = time.time()
logging.info('synthesis text {}'.format(i))
@@ -64,7 +65,7 @@ class CosyVoice:
def inference_zero_shot(self, tts_text, prompt_text, prompt_speech_16k, stream=False):
prompt_text = self.frontend.text_normalize(prompt_text, split=False)
for i in self.frontend.text_normalize(tts_text, split=True):
for i in tqdm(self.frontend.text_normalize(tts_text, split=True)):
model_input = self.frontend.frontend_zero_shot(i, prompt_text, prompt_speech_16k)
start_time = time.time()
logging.info('synthesis text {}'.format(i))
@@ -77,7 +78,7 @@ class CosyVoice:
def inference_cross_lingual(self, tts_text, prompt_speech_16k, stream=False):
if self.frontend.instruct is True:
raise ValueError('{} do not support cross_lingual inference'.format(self.model_dir))
for i in self.frontend.text_normalize(tts_text, split=True):
for i in tqdm(self.frontend.text_normalize(tts_text, split=True)):
model_input = self.frontend.frontend_cross_lingual(i, prompt_speech_16k)
start_time = time.time()
logging.info('synthesis text {}'.format(i))
@@ -91,7 +92,7 @@ class CosyVoice:
if self.frontend.instruct is False:
raise ValueError('{} do not support instruct inference'.format(self.model_dir))
instruct_text = self.frontend.text_normalize(instruct_text, split=False)
for i in self.frontend.text_normalize(tts_text, split=True):
for i in tqdm(self.frontend.text_normalize(tts_text, split=True)):
model_input = self.frontend.frontend_instruct(i, spk_id, instruct_text)
start_time = time.time()
logging.info('synthesis text {}'.format(i))